karpathy

I like to train Deep Neural Nets on large datasets.

Stanford

Github Data

Followers 102476

Following 8

Links

https://twitter.com/karpathy

AI Project

Public repos: 55Public gists: 10

nano-llama31

nanoGPT style version of Llama 3.1

star: 1.3Kfork: 73

language: Python

created at: 2024-08-01

updated at: 2025-02-09

LLM101n

LLM101n: Let's build a Storyteller

star: 32.2Kfork: 1.7K

created at: 2024-05-27

updated at: 2025-03-04

llm.c

LLM training in simple, raw C/CUDA

star: 25.9Kfork: 3.0K

language: Cuda

created at: 2024-04-08

updated at: 2025-03-04

minbpe

Minimal, clean code for the Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.

star: 9.4Kfork: 897

language: Python

created at: 2024-02-16

updated at: 2025-03-04

llama2.c

Inference Llama 2 in one file of pure C

star: 18.4Kfork: 2.2K

created at: 2023-07-23

updated at: 2025-05-08

nanoGPT

The simplest, fastest repository for training/finetuning medium-sized GPTs.

star: 39.8Kfork: 6.5K

language: Python

created at: 2022-12-28

updated at: 2025-03-04

minGPT

A minimal PyTorch re-implementation of the OpenAI GPT (Generative Pretrained Transformer) training

star: 21.5Kfork: 2.8K

language: Python

created at: 2020-08-17

updated at: 2025-03-04